The "Programmer Informatician"

The field of health and biomedical informatics is ever changing, not only just in name (noticing perhaps the recent pre-pending of “health and”), but also by the individuals who are practitioners.  Review of early works in the field show many of the seminal pieces originated from doctors who understood the impact that technology could play in healthcare [1, 2].  As the field expanded and grew, it was perhaps natural that other healthcare providers (i.e. nurses and public health practitioners) realized the special needs that pertained to their unique practice and workflow.  With this realization blossomed a growth of sub-disciplines within HBI, to the point that nursing informatics and public health informatics are vibrant components of the broader HBI community, and large enough to warrant annual meetings.

The growth, of course, is not limited to just a few disciplines within the spectrum of providing healthcare, but also to the side of those who are focused on health information technology (HIT).  There is a burgeoning growth of “programmer informaticians”, although perhaps many do not fully realize they may fall into this category.  The proliferation of HIT vendors developing systems, as well as the growth in internally developed solutions from academic and research institutions has granted a view into a class of skilled developers putting pen to paper (or more aptly, fingers to keyboards) to bring to life many novel innovations that are envisioned, designed and guided by healthcare providers involved with routine care.  The role of the “programmer informatician”, therefore, should be recognized as an organic expansion of the HBI field as well, and as such deserves some definition.

Defining the Programmer Informatician

This new role must naturally be framed within Dr. Charles Friedman’s fundamental theorem of informatics [3], which boils down to the fact that technology is not the focus of HBI.  At this point we make a classic distinction between a perhaps more classical programmer, and the programmer informatician – the former is expected to be good at developing software and focused on technology, to the point that occasionally the excitement of the technology outweighs the practicality of its implementation.  Likewise, while they may learn and understand the domain in which they work they do not dive as deeply within it.  The latter is still very much a programmer in the traditional sense, however supplemental training and experience within the field of HBI has prepared them with sufficient knowledge of the practice of healthcare that they may more judiciously apply the technology as it is appropriate.  In addition, many programmer informaticians gain experience with classical research methods and evaluation, and are able to apply these to their daily work to ensure solutions truly meet the needs of providers and patients.

Consider an example to develop software to find terms within a clinical vocabulary.  A traditional programmer can readily come up with a system to query a clinical vocabulary database that will work within the specifications.  The programmer may have additional questions that require a domain expert (i.e. “What is the structure of this vocabulary?”, “Which vocabulary do you want?”), which could extend the timeline of the project.  However, a programmer informatician may have additional training or experience with ontologies or systems like the Unified Medical Language System (UMLS) [4] and may be able to devise a more comprehensive solution.  The programmer informatician may also have the interest in evaluating their system, whether as a formal publication or just an internal operational study – a clear level outside of the unit and integration tests applied by a traditional programmer to ensure a system is working as defined.

The existence of the programmer informatician is visible today within the HBI field.  One need only do a search of software development terms in a HBI journal or PubMed to see database, software development and implementation details appearing in the literature [5-8].   However, as many programmer informaticians may be focused on application instead of classical research, much work may not be visible.

The Need for the Programmer Informatician

Within the field of HBI, software development teams exist at many institutions and are staffed by very talented programmers.  These teams have developed and will continue to develop sophisticated systems that will further clinical practice and research.  One may ask then if the traditional programmer can be sufficient, why the need exists for a programmer informatician?  While purely anecdotal, many intuitive reasons can be seen today:

Time – programmer informaticians posses a deeper understanding of the HBI landscape, which can reduce the time it takes to develop a system.  Given both training and practical experience, this understanding allows the programmer informatician to intuit an optimal solution, or at least something closer to the optimal solution.  This reduces development time (and cost!) by removing design and development iterations where a traditional programmer may head down the wrong path.

Innovation – programmer informaticians have been “bit by the informatics bug”, which means they are often times envisioning novel systems themselves.  While possibly lacking the detailed medical background as many biomedical informaticians, their enthusiasm and having the ear of other informaticians may prompt the development of new systems or the expansion of existing systems in ways not otherwise possible.

Identity – many programmers identify themselves as a “software developer within domain {X}”.  The programmer informatician identifies himself or herself within the HBI domain first, perhaps as an “informatician who is really good at programming”.  The benefits to the programmer informatician are a strong sense of identity that is recognized within the HBI community, granting them an audience in which to further their interests and careers.  Likewise, the HBI community benefits from new memberships in professional organizations as programmers find a professional home that speaks to them.

Process Improvement – studies within the software engineering and computer science domain are sometimes assumed to be domain-agnostic.  There is significant potential to evaluate if this holds true, or if specific methods could be optimized for the HBI domain.  By supplementing the existing literature with evaluations and applied descriptions of software development approaches, other programming teams may benefit just as a laboratory would from learning of a new protocol.

Thoughts for the Future

The programmer informatician exists today, whether or not they are explicitly identified as such.  The programmer informaticians themselves need to realize when they identify in this role, and should be encouraged to build a richer body of literature describing their experiences and findings.  The continued support from the HBI community will also be critical, such as through training opportunities (graduate certificate and Master programs, as well as boot camp approaches like the AMIA 10x10 program), encouragement and mentoring.  This self-actualization and support network has the potential to boost the HBI landscape as more programmer informaticians reach the field.

[1] Barnett GO. Electronic networks in hospitals. Science. 1966; 154(3756):1504.

[2] Shortliffe EH, Axline SG, Buchanan BG, Merigan TC, Cohen SN. An artificial intelligence program to advise physicians regarding antimicrobial therapy. Comput Biomed Res. 1973; 6(6):544-60.

[3] Friedman CP. A "fundamental theorem" of biomedical informatics. J Am Med Inform Assoc. 2009; 16(2):169-70.

[4] Unified Medical Language System (UMLS) – Home. National Library of Medicine.  Accessed from: http://www.nlm.nih.gov/research/umls/.  Last accessed December 22, 2011.

[5] Lenert LA, Kirsh D, Griswold WG, Buono C, Lyon J, Rao R, Chan TC. Design and evaluation of a wireless electronic health records system for field care in mass casualty settings. J Am Med Inform Assoc. 2011; 18(6):842-52.

[6] Wade TD, Hum RC, Murphy JR. A Dimensional Bus model for integrating clinical and research data. J Am Med Inform Assoc. 2011; 18(Supplement 1):i96-i102.

[7] Boyd AD, Saxman PR, Hunscher DA, Smith KA, Morris TD, Kaston M, et al. The University of Michigan Honest Broker: a Web-based service for clinical and translational research and practice. J Am Med Inform Assoc. 2009; 16(6):784-91.

[8] Huser V, Rasmussen LV, Oberg R, Starren JB. Implementation of workflow engine technology to deliver basic clinical decision support functionality. BMC Med Res Methodol. 2011; 11:43.